Consumer Prices, the Consumer Price Index, and the Cost of Living

Abstract

After presenting major findings and recommendations, the CPI Commission reiterates the estimate of a 1.1 percentage point per annum upward bias. It rejects the contention that the BLS already makes substantial corrections for quality change; that quality improvements and new products accrue only to the rich; and that procedures to make more extensive quality adjustments, valuations of new products, and adjustments for commodity and outlet substitution are impractical. The bias in the CPI can be sharply reduced, as the authors detail in this paper. Coauthors are Ellen R. Dulberger, Robert J. Gordon, Zvi Griliches, and Dale W. Jorgenson.

Estimating Real Income in the United States from 1888 to 1994: Correcting CPI Bias Using Engel Curves

“…Annual CPI bias was 0.4 percent in the 1960s and then rose to 2.7 percent between 1972 and 1982 before falling to 0.6 percent between 1982and 1994. Overall, bias between 1972and 1994 was 1.6 percent per year, in the upper end of the Boskin Commission's (Boskin et al 1998) range of 0.8-1.6 percent per year and similar to Nordhaus's (1997) and Hamilton's (2001) (Boskin et al 1998) concluded that the biggest defect in the CPI was its failure to account adequately for new goods and improvements in existing goods. Dougherty and Van Order's (1982) and Stewart and Reed's (1999) findings imply that the CPI's biggest failure has been its treatment of the homeowners housing component and that these problems were aggravated by the high inflation rates of the 1970s.…”

“…12 Because swings in housing prices are larger than swings in rents, using rents imparts a downward bias to the CPI during the housing boom in the first half of the 1920s and an upward bias from the 1925 peak to the 1935 trough in the housing market. 13 The Boskin Commission (Boskin et al 1998) estimated that the biggest source of CPI bias between 1975 and 1994 was the late introduction of new goods into the CPI and quality improvements in existing goods. The postwar pattern of higher bias in the 1970s than in the 1960s or 1980s may arise from the greater price volatility of the 1970s relative to the 1960s (Baily 1981) and from extensive improvements made to the CPI in the 1980s (see Greenlees and Mason [1996] for a full list).…”

“…The 1961 Stigler Commission concluded that virtually all economists would agree that there was upward bias in the various price indexes that they reviewed, but the commission presented no numerical estimates of bias (National Bureau of Economic Research 1961). The Boskin Commission (Boskin et al 1998) argued that CPI bias is probably greater now than it was in the past because the number of goods has grown, because a greater rate of technological change is leading to more rapid price shifts, and because demand has shifted toward services and quality, making the task of measurement much harder. However, studies of CPI bias in specific goods or commodities suggest that bias may have been greater prior to the 1950s than afterward.…”

Globalization, de-industrialization and underdevelopment in the third world before the modern era

“…Most of the latter have been for post-war United States (Gordon 1990;Boskin et al 1998) and hedonic techniques have been preferred by most investigators. Robert Lipsey and his collaborators have used the estimated impact of US product quality change on traded goods prices Lipsey 1971, 1992;Lipsey, Molinari, and Kravis 1991;Lipsey 1994).…”

International Comparisons of Trends in Economic Well-Being

“…The recent debate on 7 CPI bias is thus directly relevant to the estimation of real consumption flows. The Boskin Commission (Boskin et al, 1996) estimated that the US CPI had an upward bias of 1.1 per cent, largely due to the failure of prices indexes to capture the welfare effects of new goods and the quality improvements in existing products (Nordhaus, 1996). In this paper, no adjustment is made for potential consumer price bias.…”